Work

Bridge defect detection using small sample data with deep learning and Hyperspectral imaging

Year: 2024

Type: article

Abstract: The visual sensing method is an effective way to address long-term health monitoring of bridges. However, bridge defect detection based on visible light imaging mainly relies on grayscale and regional edge gradient information , which brings challenges such as limited information dimensions and complex background. This paper introduces a bridge defect detection method that leverages hyperspectral imaging , utilizing the unique integration of spectral and spatial information. Also a convolutional neural network algorithm with dual branches and dense blocks for spectral feature extraction is developed. This framework includes spectral and spatial branches, which independently extract respective features in order to minimize mutual interference . Compared with the support vector machine and traditional deep learning algorithms , the proposed method attains an overall model prediction accuracy(OA) of 98.57 %, an average accuracy (AA) of 98.16 %, and a Kappa coefficient of 0.9814, representing the best classification performance on small sample datasets.

Source: Automation in Construction

DOI: 10.1016/j.autcon.2024.105900

Authors: Xiong Peng, Pengtao Wang, Kun Zhou, Zhipeng Yan, Xingu Zhong, Chao Zhao

Institution: Hunan University of Science and Technology

Language: English

Topic: Infrastructure Maintenance and Monitoring

Subfield: Civil and Structural Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: hybrid

Estimated APC paid: $4,850